Signal Handling by JVM : a probable bug

I am facing the following problem. when i kill a running java code by (CTRL C SIGINT), then the sub processes created by the parent java program using "exec" call didn't get terminated. I ran the code on Solaris machine. This OS supports cascading termination meaning 'if a parent is dead then all the childs get terminated recursively". But i guess JVM calls its signal handler and didn't follow the rule of cascading termination. I THINK IT IS A BUG. can some one acknowledge/correct me on this issue

I really don't think so.
On unix boxes one can 'fork'. A fork creates a sub process (not a process.) Killing a sub process might or might not kill child sub processes.
Runtime.exec() produces a process, not a sub process. So I would never expect that killing the parent would kill the other process.
This would suggest that exiting the process would also kill the exec() process. There is a similar analog in perl. On unix it allows for fork and system/exec calls. An 'exec' terminates the perl script immediately. Using the same from the above, it would be useless since it would also immediately terminate the process just started with 'exec'.

Similar Messages

JVM Signal Handling on NT and HP-Unix

Hi
I'm trying to catch various JVM shutdown signals in my application so that the application can do some clean-up before shutting down. Here is the code that i've written...
sun.misc.SignalHandler jvmSignalHandler = new sun.misc.SignalHandler()
public void handle(sun.misc.Signal sig)
// Do some cleanup
// Call handler on interrupt signal (CTRL-C)
sun.misc.Signal.handle(new sun.misc.Signal("INT"), jvmSignalHandler);
// Call handler on terminate signal
sun.misc.Signal.handle(new sun.misc.Signal("TERM"), jvmSignalHandler);
// Call handler on abort signal
sun.misc.Signal.handle(new sun.misc.Signal("ABRT"), jvmSignalHandler);
If i run this process on NT in a dos box, the applicatin behaves fine by responding to all these signals...
e.g. If i press CTRL-C (INT), or i close the dos box (TERM), or i shutdown NT (ABRT).
However on UP-Unix, i tried kill -2 (TERM), kill -6 (ABRT) and kill -15 (INT)... Out of these, only kill -6 (ABRT) works, while the others just "kill" the application and there is no clean-up. Can anyone help if i need to use any other codes on HP-Unix, or do i need to catch any other signals in my application.
Thanks in advance
Regards
Kashif

Hope the following info(from HP-UX $man kill) is useful to you.
Signal Names and Numbers
The following table describes a few of the more common signals that
can be useful from a terminal. For a complete list and a full
description, see the header file <signal.h> and the manual entry
signal(5).
signum signame Name Description
0 SIGNULL Null Check access to pid
1 SIGHUP Hangup Terminate; can be trapped
2 SIGINT Interrupt Terminate; can be trapped
3 SIGQUIT Quit Terminate with core dump; can be trapped
9 SIGKILL Kill Forced termination; cannot be trapped
15 SIGTERM Terminate Terminate; can be trapped
24 SIGSTOP Stop Pause the process; cannot be trapped
25 SIGTSTP Terminal stop Pause the process; can be trapped
26 SIGCONT Continue Run a stopped process
BTW, do you have JavaDoc for sun.misc.Signal class in hand? I havn't. I'm not sure the name-rule for signals between HP-UX and Windows is same or not.

JVM exit handler - for jvm invoked thro' native C

complicated, atleast for me :) where do I find if JVM installs any exit handlers when invoked through native C app.
OS -> HP UX 11
java -version ->
java version "1.4.0.02"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.0.02-021031-18:00)
Java HotSpot(TM) Server VM (build 1.4 1.4.0.02-021031-19:15-PA_RISC2.0 PA2.0, mixed mode)
Although jvm used is HP's hotspot, behaviour seems to be generic, so any directions are appreciated.
In my case JNI based native shared object (on ux) is being used as a pluggin for comm with microsoft SQL2k.
A daemon loads the pluggin (shl_load) -> transaction -> unloads (shl_unload) pluggin.
For each tran I need to create a new JVM instance and later distroy it as it is not possible to multiple times create and distroy JVM in same process space (as encountered)
LD_PRELOAD=.../libjvm.sl
start the daemon( basic signals masked)
Parent -> fork() a child process and wait for its exit status.
Child -> Load JVM and execute reqd method
Child -> unload JVM
Child -> exit() the child process
Parent ->Check the child process exit() code and carry further processing.
Sometimes child does not exit cleanly. Even after exit() child process can be seen in process table and parent wait()'s idefinetly for child exit code.
gdb ouput of child preocess
#0 0xdd28ca38 in os::report_fatal_error () from /opt/java1.4/jre/lib/PA_RISC2.0/server/libjvm.sl
#1 0xdd28dc48 in os::handle_unexpected_exception ()
from /opt/java1.4/jre/lib/PA_RISC2.0/server/libjvm.sl
#2 0xdd296c50 in os::Hpux::JVM_handle_hpux_signal ()
from /opt/java1.4/jre/lib/PA_RISC2.0/server/libjvm.sl
#3 0xdd293adc in os::Hpux::signalHandler ()
from /opt/java1.4/jre/lib/PA_RISC2.0/server/libjvm.sl
#4 <signal handler called>
#5 0x7e5e6134 in ?? ()
#6 0xc03c6fbc in __niamHelper () from /usr/lib/libCsup.2
Can any one tell me what exactly is happening ?
If I call [b]_exit() instead of exit() everything works fine.

Thnx a lot,
Finally I was able to locate atexit()'ed function which was causing the problem (setting br at exit() was not working, execution straightway stopped at #0)
(gdb) thr 1
[Switching to thread 1 (system thread 17555)]
(gdb) bt
#0 0x7e6224e0 in epc_exit_handler () from /citisafe/users/yogiraj/cs35/dll/ora_tcp.sl
#1 0xc03c6fbc in __niamHelper () from /usr/lib/libCsup.2
#2 0xc03c7110 in niambody () from /usr/lib/libCsup.2
#3 0xc03c721c in _niam () from /usr/lib/libCsup.2
#4 0xc016fe20 in exit () from /usr/lib/libc.2
#5 0xc4f0c364 in sql_exec_custStrProc () from /citisafe/users/yogiraj/cs35/lib/libtcsql.sl
#6 0x7ec0c9f4 in exec_ext_mgmt (p_ExtCredentials=0x7edf2878)
at /citisafe/users/yogiraj/cs35/c/sql_app.c:251
#7 0x9620 in adm_process_rec ()
#8 0x8bc8 in process_pwd_requests ()
#9 0x8684 in adm_do_ext_sync ()
#10 0x7f5c in main ()
Oracle client lib was the culprit. Daemon is basically a 'C' daemon which loads pluggins for communication with various DB's including Oracle, Sybase, MS SQL etc and later unloads them on completion.
Ora lib libclntsh.sl installed "epc_exit_handler", eventhough we had shl_unload()'ed oracle pluggin earlier the handler refrence was still around. Later when process exit()'ed may be static dealloaction mech panicked.
Above exit() handler execution even after unloading of Oracle libs seems to be a known bug and work around suggested is to LD_PRELOAD=%ORALIB/libclntsh.sl. After LD_PRELOAD'ing exit() works fine.

[SOLVED] fglrx(0): [drm] failed to remove DRM signal handler

Hello!
I'm new to Arch so please forgive me if I'm asking about something obvious.
After spending many hours on reading and comprehension of Arch philosophy I finally dared to install Arch 2007.08-02 (Don't Panic) on my computer Fortunately I was prepared for text-mode configuration, but obviously not prepared enough... My first major problem is with the X server. I have followed strictly the instructions from the wiki and installed xorg and catalyst drivers afterwards (for Radeon X800 Pro). Now probably something is wrong with my xorg.conf, which you can see below, but there's also a bit more sophisticated problem causing this type of message in /var/log/Xorg.0.log
(II) fglrx(0): Shutdown CMMQS
(EE) fglrx(0): [drm] failed to remove DRM signal handler
(II) fglrx(0): [drm] removed 1 reserved context for kernel
(II) fglrx(0): [drm] unmapping 8192 bytes of SAREA 0x2000 at 0xb7c78000
(II) fglrx(0): [drm] Closed DRM master.
I couldn't find a solution either on this board and on the web as well, so I dare to ask for a little tip what to do to solve this
Thx in advance.
P.S. My xorg.conf
# Auto-generated by Archie mkxcfg
Section "ServerLayout"
Identifier "My Layout"
Screen 0 "aticonfig-Screen[0]" 0 0
InputDevice "Keyboard0" "CoreKeyboard"
InputDevice "Mouse1" "CorePointer"
EndSection
Section "Files"
# FontPath "/usr/share/fonts/75dpi:unscaled"
# FontPath "/usr/share/fonts/75dpi"
# Additional fonts: Locale, Gimp, TTF...
# FontPath "/usr/share/lib/X11/fonts/latin2/75dpi"
# FontPath "/usr/share/lib/X11/fonts/latin2/100dpi"
# True type and type1 fonts are also handled via xftlib, see /etc/X11/XftConfig!
# FontPath "/usr/share/fonts/Type1"
RgbPath "/usr/share/X11/rgb"
ModulePath "/usr/lib/xorg/modules"
FontPath "/usr/share/fonts/misc:unscaled"
FontPath "/usr/share/fonts/misc"
FontPath "/usr/share/fonts/100dpi:unscaled"
FontPath "/usr/share/fonts/100dpi"
FontPath "/usr/share/fonts/PEX"
FontPath "/usr/share/fonts/cyrillic"
FontPath "/usr/share/fonts/ttf/western"
FontPath "/usr/share/fonts/ttf/decoratives"
FontPath "/usr/share/fonts/truetype"
FontPath "/usr/share/fonts/truetype/openoffice"
FontPath "/usr/share/fonts/truetype/ttf-bitstream-vera"
FontPath "/usr/share/fonts/latex-ttf-fonts"
FontPath "/usr/share/fonts/defoma/CID"
FontPath "/usr/share/fonts/defoma/TrueType"
EndSection
Section "Module"
# Load "ddc" # ddc probing of monitor
# Load "type1"
# Load "synaptics"
Load "dbe"
Load "dri"
Load "extmod"
Load "glx"
Load "bitmap" # bitmap-fonts
Load "freetype"
# Load "record"
EndSection
Section "ServerFlags"
Option "AllowMouseOpenFail" "true"
EndSection
Section "InputDevice"
Identifier "Keyboard0"
Driver "kbd"
Option "XkbModel" "logiaccess"
Option "XkbLayout" "pl"
Option "XkbOptions" "compose:rwin"
EndSection
Section "InputDevice"
Identifier "Mouse1"
Driver "mouse"
Option "Protocol" "ExplorerPS/2"
Option "Device" "/dev/input/mouse1"
Option "Buttons" "7"
Option "ZAxisMapping" "4 5"
Option "ButtonMapping" "1 2 3 6 7"
EndSection
Section "Monitor"
Identifier "aticonfig-Monitor[0]"
Option "VendorName" "ATI Proprietary Driver"
Option "ModelName" "Generic Autodetecting Monitor"
Option "DPMS" "true"
EndSection
Section "Device"
Identifier "aticonfig-Device[0]"
Driver "fglrx"
EndSection
Section "Screen"
Identifier "aticonfig-Screen[0]"
Device "aticonfig-Device[0]"
Monitor "aticonfig-Monitor[0]"
DefaultDepth 24
SubSection "Display"
Viewport 0 0
Depth 24
EndSubSection
EndSection
Section "DRI"
Mode 0666
EndSection
Last edited by Zibi1981 (2008-02-20 10:49:20)

Hello quad3d@work Thanks for the interest, but I've already solved this problem. I think it was something with my xorg.conf file, because when I made some changes to it, suddenly all started to work. If You are still interested in the output of the above command, well, here it is
[zibi1981@felipe ~]$ cat ~/.xinitrc
#!/bin/sh
# ~/.xinitrc
# Executed by startx (run your window manager from here)
# exec gnome-session
# exec startkde
# exec startxfce4
# exec icewm
# exec openbox
# exec blackbox
# exec fluxbox
# exec ratpoison
# exec dwm
# ... or any other WM of your choosing ...
# exec xterm
All the lines are commented, because I use Entrance with appropriate input in /etc/inittab file
[zibi1981@felipe ~]$ cat /etc/inittab
# /etc/inittab
# Runlevels:
# 0 Halt
# 1(S) Single-user
# 2 Not used
# 3 Multi-user
# 4 Not used
# 5 X11
# 6 Reboot
## Only one of the following two lines can be uncommented!
# Boot to console
#id:3:initdefault:
# Boot to X11
id:5:initdefault:
rc::sysinit:/etc/rc.sysinit
rs:S1:wait:/etc/rc.single
rm:2345:wait:/etc/rc.multi
rh:06:wait:/etc/rc.shutdown
su:S:wait:/sbin/sulogin -p
# -8 options fixes umlauts problem on login
c1:2345:respawn:/sbin/agetty -8 38400 vc/1 linux
c2:2345:respawn:/sbin/agetty -8 38400 vc/2 linux
c3:2345:respawn:/sbin/agetty -8 38400 vc/3 linux
c4:2345:respawn:/sbin/agetty -8 38400 vc/4 linux
c5:2345:respawn:/sbin/agetty -8 38400 vc/5 linux
c6:2345:respawn:/sbin/agetty -8 38400 vc/6 linux
ca::ctrlaltdel:/sbin/shutdown -t3 -r now
# Example lines for starting a login manager
#x:5:respawn:/usr/bin/xdm -nodaemon
#x:5:respawn:/usr/sbin/gdm -nodaemon
#x:5:respawn:/opt/kde/bin/kdm -nodaemon
#x:5:respawn:/usr/bin/slim >& /dev/null
x:5:respawn:/usr/sbin/entranced -nodaemon >& /dev/null
# End of file
Anyway, thanks for the interest once more

OS Signal Handling problem

Hi, I have a problem trying to handle operating system signals. The exception I get is: Exception already handled by de JVM.
The enviroment that runs my application is:
-IBM AIX 4.3
-IBM JRE 1.3.0
The process is launched with "nohup"
The complete line is: nohup java -jar -Dlog4j.configuration=log.xml etec.jar >/dev/null 2>/dev/null &The code is:
try
if (senyalInterrupcion != null)
    Signal.handle(new Signal (signalName), new SignalHandler ()
      public void handle(Signal arg0)
        System.exit(0);
catch (IllegalArgumentException ex)
ex.printStackTrace ();
}I know that the use of com.sun.* classes is not recommended by Sun, but I need to communicate the JVM in wich my application run with the shell, if anyone knows a better method (except RMI) please let me know.
P.S. Sorry for my english
Thanks in advance,
Juanjo

As of jdk1.3, run w/ Java -Xrs option (i.e., java -Xrs myProg) so default JVM signal handlers are not run. Then, you can install your own signal handlers using sun.misc.Signal and SignalHandler and they will not conflict with the JVM's handlers. See the following article for excellent coverage of the topic:
http://www-106.ibm.com/developerworks/ibm/library/i-signalhandling/

64-bit sigaction() Signal Handler for aio_write()

I'm having trouble using a sigaction() specified signal handler when built as a 64 bit executable (-xarch=v9).
The siginfo_t argument gets passed to the signal handler just fine. The problem is with the si_value.sival_ptr member. The other members (sig, type, pid ) are all okay.
When built for 32-bits everything works fine.
When built for 64-bits all I get is the high 32-bits of the sival_ptr, the low 32-bits are always 0.
The signal is being generated as a result of a call to aoi_write().
Please help - I'm in a hurry and I can't seem to find the answer.
I can do a work-around using sival_int (it's in union with sival_ptr) and indexing a table for an address, but that's a hassle, while the pointer method is clean.
Any help will be appreciated.
Bruce.

Be careful that the program is built at compile time
with the correct 64-bit library and not inadvertantly
with a 32-bit version. The fact that the failure is
a seg-fault and dbx can't resolve the symbol make the
system lib a suspect and the problem a 32 bit vs.
64 bit version.

Unix Signal Handling in Oracle Apps

I have a number of questions regarding Signal Handling
on Unix platforms within Oracle client code. I would like
to ensure that my signal handling requirements do not clash with
those of Oracle within an application.
1. Does the Database client code register any Unix signal
handling routines within a user application?
2. If it does, which signals does it handle?
3. What is the scope of these handler routines, i.e. when are
they registered and when do
they become de-registered?
4. When registered, do these handler routines attempt to pass
signals on to previously registered
handlers (i.e. do previously registered handlers get chained)?
5. When they are de-registered do they attempt to re-instate any
previously registered signal handlers?
thank you in advance for your help,
John Tresadern
IBM UK.
null

This can certainly be done, assuming you have sufficient hardware/storage resources.
Each instance will need to have a separate Unix account (e.g. applvis, appldev etc).
The actual details of the cloning process are outlined in MOS Doc 230672.1 (Cloning Oracle Applications Release 11i with Rapid Clone)
HTH
Srini

OCCI & Signal Handling

Hello,
Before calling any OCCI function my signal handling (ie for SIGTERM) works, however, after calling OCCI environment/connection my signal handling no longer works. I believe the OCCI version is 9.2.0.6.0. Does anyone have an example of using signals with OCCI? It seems OCCI is 'stealing' my signals! Below is the basic test program.
Thankyou in advance!
Ross.
The problem area: (c++)
oracle::occi::Environment *env;
oracle::occi::Connection *conn;
oracle::occi::Statement *stmt;
oracle::occi::ResultSet *rset;
::signal(SIGINT, signal_handle);
::signal(SIGTERM, signal_handle);
::signal(SIGPIPE, sigpipe_handle);
cout << "Attempting to connect to server" << endl;
try
env = oracle::occi::Environment::createEnvironment (oracle::occi::Environment::DEFAULT);
conn = env->createConnection ("usr", "pswd", "db");
cout << "Connection Estabolished OK" << endl;
string sqlStmt = "SELECT I_STATUS FROM K_INC";
stmt = conn->createStatement (sqlStmt);
rset = stmt->executeQuery ();
while (rset->next ())
cout << "Result!" << rset->getString(1) << endl;
stmt->closeResultSet (rset);
conn->terminateStatement (stmt);
catch(oracle::occi::SQLException ex)
cout<<"Exception thrown for displayAllRows"<<endl;
cout<<"Error number: "<< ex.getErrorCode() << endl;
cout<<ex.getMessage() << endl;
cout << "We have a problem!" << endl;
}

hi Ross,
I think there is no problem of signal stealing . I am able to use the signal handling before the connection, after the connection and even after insertion of rows into the table.
I implicitly raised a signal in all the above three cases and also explicitly from another process using kill -15 <process ID >, and the signal handler seems to work fine.
the code is attached below , see if it helps.
thanks
varun
#include <iostream>
#include <occi.h>
#include<csignal>
using namespace oracle::occi;
using namespace std;
int i=1;
class occidml
private:
Environment *env;
Connection *conn;
Statement *stmt;
public:
occidml (string user, string passwd, string db)
env = Environment::createEnvironment (Environment::DEFAULT);
conn = env->createConnection (user, passwd, db);
~occidml ()
env->terminateConnection (conn);
Environment::terminateEnvironment (env);
* Inserting a row into the table.
void insertRow ()
string sqlStmt = "INSERT INTO varun VALUES (111, 'ASHOK')";
stmt = conn->createStatement (sqlStmt);
try{
stmt->executeUpdate ();
cout << "insert - Success" << endl;
}catch(SQLException ex)
cout<<"Exception thrown for insertRow"<<endl;
cout<<"Error number: "<< ex.getErrorCode() << endl;
cout<<ex.getMessage() << endl;
conn->terminateStatement (stmt);
void term(int sig)
cout << "handling signal no." << sig << endl;
i=i+1;
int main (void)
string user = "SCOTT";
string passwd = "TIGER";
string db = "";
signal(SIGTERM,term); // register a SIGTERM handler
while(i==1){}
printf("Before Connection\n");
occidml *demo = new occidml (user, passwd, db);
while (i==2){}
printf("After Connection\n");
cout << "Inserting a record into the table author_tab " <<endl;
demo->insertRow ();
while(i==3){}
delete(demo);
printf("Complete");
}

Signal handler in Tuxedo

Hi,
We are using Tuxedo 8.0 in our application.
Our servers are crashing due to some code in our app...
We want to use C signal handler to trap this signals to trace the code, before crashing. This way we can easily fix the code.
But for some reason when I use signal, its not catching SIGBUS. I also used Usignal, but still to no success...
Please suggest...
Thanks
Rama

Hi,
I have not encountered any problem with signal trapping with Tuxedo 6.5 and 8.1, i have not worked on 8.0. May i know how you are using signals in your service. In our case we are using as --
service A()
signal(SIGBUS , SigCatcher) ;
signal(SIGSEGV , SigCatcher) ;
signal(SIGINT , SigCatcher) ;
signal(SIGTERM , SigCatcher) ;
signal(SIGQUIT , SigCatcher) ;
signal(_SIGKILL , SigCatcher) ;
int SigCatcher(int signo)
/* CODE to Handle signal *****************/
Regds
-Ashish

C5 - Two probably bugs (sms & white screen)

I have a C5-00 it works very well and I really appreciate it. I have update the firmware to 065.001.
But I found two probably bugs:
1) when I open a received sms, then go to the previous one moving with the "joypad" (i.e. without returning to the list of received sms) and then return to the first one sms the last two lines of the sms are disappeared! I have to exit to the list of sms' then re-open the sms to see the sms entirely, i.e with all the lines.
2) The battery recharger is attached to the phone. Then the alarm clock of the phone rings and I shut down the ring. I try to switch on the phone: the monitor became completely white (neither the symbol of the charged battery appear). I have to disconnect the battery to switch on again the phone.
Do somebody experienced the same issues?
I hope Nokia can explain why they happen and, if the case, correct this issues/bugs.
Thanks,
alemanowar

Hi, Exactly the same bug here. A possible workaround is to switch between two SMS or scroll down to see the whole text. @vijayt It IS a bug, because it worked well with 032.xxx. I dont wont to scroll down to see the text - the display is big enough to show the whole message. I will collect here the other posts for the same topic: * /t5/Nseries-and-S60-Smartphones/6700-slide-truncated-sms-after-upgrade/m-p/937351/highlight/true#M25... * /t5/Cseries/Nokia-C5-Texts-are-half/m-p/934725/highlight/true#M13715 So I hope this will be fixed in 062.005, which is still not available for my phone :-( Is there a "real" bugtracking system one (not developer) can use? Regards, Christoph

Signal Handling

Friends,
The scenario is such that, a server is running on several ports satisfying multiple clients on each of the port.
The requirement is that the server must catch signals like killing of the server or restarting of the server.
Will I code the signal handler by adding a shutdown hook to the Java runtime, using the
Runtime.getRunTime ().addShutdownHook method.
If I kill the process by kill -9 option will it serve the purpose.
Please give a reply its urgent

Hi ....
Actually is there any class in java for Signal Handling which can handle the following signals :
SIGCHLD
SIGHUP
SIGTERM
SIGINT
SIGBUS
SIGSEGV
SIGPIPE
I would run the program in Suse Linux Environment.... If I want to avoid JNI code ..... then what is the other option ......
Please reply....

Signal Handling on Alpha machines

Could any one of you tell me how to trap 'Ctl-C' on Alpha machines or OSF1 OS ? The following piece of code works fine on Linux and Solaris bt not on alphas ?
I know sun.misc.* is not not guranteed to work on all platforms but wanted to see if any of you ran into this problem and has a solution?
Thanks
Sreenath
import sun.misc.Signal;
import sun.misc.SignalHandler;
public class InterruptSignalHandler implements SignalHandler {
public static boolean USER_INTERRUPTION = false;
private String SIG_INT = "INT";
private String SIG_TERM = "TERM";
private Signal intSignal = null;
private Signal termSignal = null;
public InterruptSignalHandler() {
public void install() {
intSignal = new Signal(SIG_INT);
termSignal = new Signal(SIG_TERM);
Signal.handle(intSignal, this);
Signal.handle(termSignal, this);
* The <code>handle</code> method here.
* @param sig a <code>Signal</code> value
public void handle(Signal sig) {
System.out.println("Tried to kill me...");
if (USER_INTERRUPTION == false) {
USER_INTERRUPTION = true;
System.out.println("I am interrupted.."+USER_INTERRUPTION);
* The <code>main</code> method here.
* @param args[] a <code>String</code> value
public static void main(String args[]) {
System.out.println("Try to destroy me by pressing CTRL-C");
InterruptSignalHandler intSigHandler = new InterruptSignalHandler();
intSigHandler.install();
// Start infinite loop
for (;;){
System.out.print(".");
}

Wasn't it enough to simply reply to your other (duplicate) question you posted yesterday??? Does my answer have no relevancy whatsoever???
http://forum.java.sun.com/thread.jsp?thread=511180&forum=31&message=2429068

Signal Handling Issue with OCI

Greetings.
I have a library using OCI with Oracle 10g.
I am not doing any signal handling inside the library.
I have realized after a connection is made to Oracle 10g server and successfully terminated, the ctrl+Z signal (SIGTSTP) would not be honored. If I use the same library and make no connection to database, then the signal is honored. Other signals such as SIGINT etc are OK even if I make connections to database server.
Operations that I do using OCI are simple:
1- initialization of context (standard env creation, set attributes, handle allocation etc)
2- making connection
3- inserting using SQL insert method (statement prep, defineByPos, execute)
4- simple selection of last value of a sequence (statement prep, defineByPos, execute)
5- clean up (OCISessionEnd, OCIServerDetach,and freeing handles)
Again, I have no signal handling inside the library. I am using this over Solaris OS and the functionality is correct except for this signal issue.
My questions:
1- Is this a known issue or is this something that has been seen before?
2- what sort of signal handling is in OCI that could potentially impact my app?
3- Is there any solution that you could suggest?
thank you in advance.

I'm trying to migrate from oracle 9i HP unix to oracle 10G AIX5.3
I'm trying to create a connection pool. Then create Many threads, with each of these
performing an OCI Session Get at thread initialization.
Then handing the threads work to perform on their existing OCI session to the connectionpool.
Same thing was working fine in oracle 9i but if I try to do the same in oracle 10G seems to get locked up,
unless the sessionget/sessionrelease are in the same working
thread subroutine. I am trying to avoid getting a session and releasing for each unit of
work.Instead I'd like to acquire a bunch of session against the connection pool.And then have the threads (with persistent sessions) perform periodic work....
If utilizing a get/work/release all in the same thread and subroutine call.Then everything works, except performance is very poor.
please find the function I'm using to achive the same.
Experts please help me ,I'm totally struck here....
int ConnectionOpen (Connection conn, char username, char password, char server) {
** Functionality : Opens a new session with Oracle.
** Parameters : conn - pointer to the struct Connection. It is an output parameter
** returns the handle to the new session.
** username - pointer to the char array has username.
** password - pointer to the char array has password.
** server - pointer to the char array has database name.
** Returns 0 on success, 1 on success with information and non-zero value on error.
int status = 0;
boolean sesfnd;
text errbuf[512];
sb4 errcode = 0;
if (!conn) return OCI_INVALID_HANDLE;
if (!strcasecmp (USE_SESS_POOL, "YES"))
/* PLC44050021 Session Pool Changes*/
if (conopen==0)
status = OCIEnvCreate((OCIEnv **) &envhp, OCI_THREADED,
(dvoid *)0, (dvoid *(*)(dvoid *, size_t))0,
(dvoid *(*)(dvoid *, dvoid *, size_t))0, (void (*)(dvoid *, dvoid *))0,
(size_t)0, (dvoid**)0);
if (status != 0 && status != 1) return status; /*error in creating environment*/
status = OCIHandleAlloc((dvoid *)envhp,
(dvoid **)&errhp,
OCI_HTYPE_ERROR,
(size_t)0,
(dvoid **)0);
if ( status != 0 && status != 1 ) return status; /*error in allocating error handler*/
status = OCIHandleAlloc((dvoid *)envhp,
(dvoid **)&authp,
OCI_HTYPE_AUTHINFO,
(size_t)0,
(dvoid **)0);
if ( status != 0 && status != 1 ) return status; /*error in allocating error handler*/
status = OCIHandleAlloc((dvoid *)envhp,
(dvoid **) &poolhp,
OCI_HTYPE_SPOOL,
(size_t)0,
(dvoid **)0);
if ( status != 0 && status != 1 ) return status; /*error in allocating CPOOL Handler*/
status = OCISessionPoolCreate((OCIEnv*)envhp,
(OCIError*)errhp,poolhp,
&poolName, (ub4*)&poolNameLen,
(text*)server, (ub4)strlen(server),
sessMin, sessMax, sessIncr,
(text*)username, (ub4)strlen(username),
(text*)password, (ub4)strlen(password),
OCI_DEFAULT) ;
if ( status != 0 && status != 1 ) return status;
conn->envhp=(OCIEnv*)envhp;
conn->errhp=(OCIError*)errhp;
status=OCIAttrSet((dvoid *)authp,(ub4) OCI_HTYPE_AUTHINFO,
(text*)username, (ub4)strlen(username),
(ub4) OCI_ATTR_USERNAME, (OCIError*)conn->errhp);
if ( status != 0 && status != 1 ) return status;
status=OCIAttrSet((dvoid *) authp,(ub4) OCI_HTYPE_AUTHINFO,
(text*)password, (ub4)strlen(password),
(ub4) OCI_ATTR_PASSWORD, (OCIError*)conn->errhp);
if ( status != 0 && status != 1 ) return status;
status = OCISessionGet((OCIEnv*)conn->envhp,
(OCIError*)conn->errhp,
(OCISvcCtx**)&(conn->svchp),
authp,
(OraText *)poolName, (ub4)strlen((char*)poolName),
(OraText *)"", (ub4)0,
NULL, NULL,
0,
OCI_SESSGET_SPOOL);
if ( status != 0 && status != 1 ) return status;
conopen=1;
/******* if the session pool is already established, reuse sessions out of it ********** /
else
conn->errhp=(OCIError*)errhp;
conn->envhp=(OCIEnv*)envhp;
/******* Hanging in this call ********** /
status = OCISessionGet((OCIEnv*)conn->envhp,
(OCIError*)conn->errhp,
(OCISvcCtx**)&(conn->svchp),
authp,
(OraText *)poolName, (ub4)strlen((char*)poolName),
(OraText *)"", (ub4)0,
NULL, NULL,
0,
OCI_SESSGET_SPOOL);
if ( status != 0 && status != 1 ) return status;
else
status = OCIEnvCreate((OCIEnv **) &conn->envhp, (ub4)OCI_DEFAULT,
(dvoid *)0, (dvoid *(*)(dvoid *, size_t))0,
(dvoid *(*)(dvoid *, dvoid *, size_t))0, (void (*)(dvoid *, dvoid *))0,
(size_t)0, (dvoid**)0);
if (status != 0 && status != 1) return status; /*error in creating environment*/
status = OCIHandleAlloc((dvoid *)conn->envhp, (dvoid **)&conn->errhp, OCI_HTYPE_ERROR, (size_t)0, (dvoid **)0);
if ( status != 0 && status != 1 ) return status; /*error in allocating error handler*/
status = OCILogon((OCIEnv*)conn->envhp, (OCIError*)conn->errhp, (OCISvcCtx**)&(conn->svchp),
(text*)username, (ub4)strlen(username),
(text*)password, (ub4)strlen(password),
(text*)server, (ub4)strlen(server));
return status;
}

Signal Handler

hi all.
I encountered a very strange problem recently on Solaris 8 and forte C++ 6.2.
#include "signal.h"
extern "C" sig_handler(int)
// do something
if (signal(SIGUSR1, sig_handler) == SIG_ERR) {
cerr << "Errors aborting!" << endl;
else{
cout << "OK" << endl;
First, it is compiled and gives me OK when I start the program.
but when SIGUSR1 is used, nothing happens.
I change signal(SIGUSR1, sig_handler) to sigset(SIGUSR1, sig_handler) ,
then it is working.
Any idea?
Thanks
CC

So what? How should iPlanet or any other product be able to change signal handlers in other processes? That's simply impossible.
When using the old (obsolete?) signal() call the signal handler must reestablish itself via signal() after it got called.
signal() has severe problems that's why sigaction() was introduced. I'd recommend to use sigaction()...

Probable bug in Kodo class handling code (Beta 2.2)

I have a test class "Widget" that has two attributes, name and
mark. I am encountering a curious error in the no-args
constructor under some circumstances that looks like a bug.
public class Widget
private static Random random = new Random();
private String name;
private int mark;
private Widget()
public Widget(String name)
this.name = name;
mark = random.nextInt(1000);
Typically, the test application constructs Widgets with a name.
But since I believe that it is good practice for the application
to define the no-args constructor even when it is only for the
use of JDO, I have a private no-args constructor. As the class
stands above, it compiles, enhances, and runs just fine.
But if I copy the line "mark = random.nextInt(1000);" into the
no-args constructor, the code compiles and enhances just fine,
but bombs on running with the following error:
javax.jdo.JDOFatalDataStoreException:
The registered class "com.ysoft.jdo.book.widget.Widget"
is not compiled or not longer exists. If the class has been deleted,
unregister it before proceeding.
NestedThrowables:
java.lang.ExceptionInInitializerError
     at
com.solarmetric.kodo.impl.jdbc.schema.DB.getPersistentTypes(DB.java:270)
     at com.solarmetric.kodo.impl.jdbc.JDBCPersistenceManagerFactory.setup
(JDBCPersistenceManagerFactory.java:170)
     at
com.solarmetric.kodo.runtime.PersistenceManagerFactoryImpl.privateSetup
(PersistenceManagerFactoryImpl.java:501)
     at
com.solarmetric.kodo.runtime.PersistenceManagerFactoryImpl.getPersistenceManager
(PersistenceManagerFactoryImpl.java:61)
     at
com.solarmetric.kodo.runtime.PersistenceManagerFactoryImpl.getPersistenceManager
(PersistenceManagerFactoryImpl.java:50)
     at
com.ysoft.jdo.book.widget.WidgetHandler.<init>(WidgetHandler.java:43)
     at com.ysoft.jdo.book.test.widget.TestWidget.main(TestWidget.java:18)
Exception in thread "main"
Using the schematool for -action refresh will yield a null pointer
exception at the no-args constructor line that is setting the
value of mark.
After initially having the mark initialization code in the
no-args constructor and puzzling over the error, I realized that
it didn't make sense to set the value of mark in the no-args
constructor, since JDO would later set it to whatever the value
in the database was. But it seems to me that there should be no
great harm in it, either. If mark is initialized with the value
"42" instead of "random.nextInt(1000)", there are no problems.
Any insights?
David Ezzio
Yankee Software

David,
I'm not quite sure what to do about this. I'll have to dig into the spec
and get back to you.
Here's what's happening:
Upon compilation, your initialization of the static field 'random' is
inserted into a 'static' block.
Upon bytecode enhancement, a good deal of code is added to the 'static'
block. This code is responsible for registering the class with the JDO
system.
One step of this registration process involves instantiating an object
of type 'Widget' and passing it to JDOImplHelper.registerClass().
This step is inserted into the static block before the initialization
of 'random'.
So, when the no-args constructor is invoked in preparation for the call
to registerClass(), 'random' is still null. Hence the null pointer
instruction.
As a temporary workaround, move the 'random' initialization code into an
init () method that is invoked from the constructors, and which
initializes 'random' if it is null. Or just don't perform the
initialization in the no-args constructor.
-Patrick Linskey

Signal Handling by JVM : a probable bug

Similar Messages

Maybe you are looking for